1
|
Pinto J, Ramos JRC, Costa RS, Rossell S, Dumas P, Oliveira R. Hybrid deep modeling of a CHO-K1 fed-batch process: combining first-principles with deep neural networks. Front Bioeng Biotechnol 2023; 11:1237963. [PMID: 37744245 PMCID: PMC10515724 DOI: 10.3389/fbioe.2023.1237963] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/10/2023] [Accepted: 08/22/2023] [Indexed: 09/26/2023] Open
Abstract
Introduction: Hybrid modeling combining First-Principles with machine learning is becoming a pivotal methodology for Biopharma 4.0 enactment. Chinese Hamster Ovary (CHO) cells, being the workhorse for industrial glycoproteins production, have been the object of several hybrid modeling studies. Most previous studies pursued a shallow hybrid modeling approach based on three-layered Feedforward Neural Networks (FFNNs) combined with macroscopic material balance equations. Only recently, the hybrid modeling field is incorporating deep learning into its framework with significant gains in descriptive and predictive power. Methods: This study compares, for the first time, deep and shallow hybrid modeling in a CHO process development context. Data of 24 fed-batch cultivations of a CHO-K1 cell line expressing a target glycoprotein, comprising 30 measured state variables over time, were used to compare both methodologies. Hybrid models with varying FFNN depths (3-5 layers) were systematically compared using two training methodologies. The classical training is based on the Levenberg-Marquardt algorithm, indirect sensitivity equations and cross-validation. The deep learning is based on the Adaptive Moment Estimation Method (ADAM), stochastic regularization and semidirect sensitivity equations. Results and conclusion: The results point to a systematic generalization improvement of deep hybrid models over shallow hybrid models. Overall, the training and testing errors decreased by 14.0% and 23.6% respectively when applying the deep methodology. The Central Processing Unit (CPU) time for training the deep hybrid model increased by 31.6% mainly due to the higher FFNN complexity. The final deep hybrid model is shown to predict the dynamics of the 30 state variables within the error bounds in every test experiment. Notably, the deep hybrid model could predict the metabolic shifts in key metabolites (e.g., lactate, ammonium, glutamine and glutamate) in the test experiments. We expect deep hybrid modeling to accelerate the deployment of high-fidelity digital twins in the biopharma sector in the near future.
Collapse
Affiliation(s)
- José Pinto
- LAQV-REQUIMTE, Department of Chemistry, NOVA School of Science and Technology, NOVA University Lisbon, Caparica, Portugal
| | - João R. C. Ramos
- LAQV-REQUIMTE, Department of Chemistry, NOVA School of Science and Technology, NOVA University Lisbon, Caparica, Portugal
| | - Rafael S. Costa
- LAQV-REQUIMTE, Department of Chemistry, NOVA School of Science and Technology, NOVA University Lisbon, Caparica, Portugal
| | | | | | - Rui Oliveira
- LAQV-REQUIMTE, Department of Chemistry, NOVA School of Science and Technology, NOVA University Lisbon, Caparica, Portugal
| |
Collapse
|
2
|
Pandey AK, Park J, Ko J, Joo HH, Raj T, Singh LK, Singh N, Kim SH. Machine learning in fermentative biohydrogen production: Advantages, challenges, and applications. BIORESOURCE TECHNOLOGY 2023; 370:128502. [PMID: 36535617 DOI: 10.1016/j.biortech.2022.128502] [Citation(s) in RCA: 3] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 10/31/2022] [Revised: 12/11/2022] [Accepted: 12/14/2022] [Indexed: 06/17/2023]
Abstract
Hydrogen can be produced in an environmentally friendly manner through biological processes using a variety of organic waste and biomass as feedstock. However, the complexity of biological processes limits their predictability and reliability, which hinders the scale-up and dissemination. This article reviews contemporary research and perspectives on the application of machine learning in biohydrogen production technology. Several machine learning algorithems have recently been implemented for modeling the nonlinear and complex relationships among operational and performance parameters in biohydrogen production as well as predicting the process performance and microbial population dynamics. Reinforced machine learning methods exhibited precise state prediction and retrieved the underlying kinetics effectively. Machine-learning based prediction was also improved by using microbial sequencing data as input parameters. Further research on machine learning could be instrumental in designing a process control tool to maintain reliable hydrogen production performance and identify connection between the process performance and the microbial population.
Collapse
Affiliation(s)
- Ashutosh Kumar Pandey
- Department of Civil and Environmental Engineering, Yonsei University, Seoul 03722, Republic of Korea
| | - Jungsu Park
- Department of Civil and Environmental Engineering, Yonsei University, Seoul 03722, Republic of Korea
| | - Jeun Ko
- Department of Civil and Environmental Engineering, Yonsei University, Seoul 03722, Republic of Korea
| | - Hwan-Hong Joo
- Department of Civil and Environmental Engineering, Yonsei University, Seoul 03722, Republic of Korea
| | - Tirath Raj
- Department of Civil and Environmental Engineering, Yonsei University, Seoul 03722, Republic of Korea
| | - Lalit Kumar Singh
- Department of Biochemical Engineering, Harcourt Butler Technical University, Kanpur 208002, Uttar Pradesh (UP), India
| | - Noopur Singh
- Dr. A. P. J. Abdul Kalam Technical University, Lucknow, Uttar Pradesh (UP), India
| | - Sang-Hyoun Kim
- Department of Civil and Environmental Engineering, Yonsei University, Seoul 03722, Republic of Korea.
| |
Collapse
|
7
|
Lee D, Jayaraman A, Kwon JS. Development of a hybrid model for a partially known intracellular signaling pathway through correction term estimation and neural network modeling. PLoS Comput Biol 2020; 16:e1008472. [PMID: 33315899 PMCID: PMC7769624 DOI: 10.1371/journal.pcbi.1008472] [Citation(s) in RCA: 20] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/23/2020] [Revised: 12/28/2020] [Accepted: 10/26/2020] [Indexed: 12/30/2022] Open
Abstract
Developing an accurate first-principle model is an important step in employing systems biology approaches to analyze an intracellular signaling pathway. However, an accurate first-principle model is difficult to be developed since it requires in-depth mechanistic understandings of the signaling pathway. Since underlying mechanisms such as the reaction network structure are not fully understood, significant discrepancy exists between predicted and actual signaling dynamics. Motivated by these considerations, this work proposes a hybrid modeling approach that combines a first-principle model and an artificial neural network (ANN) model so that predictions of the hybrid model surpass those of the original model. First, the proposed approach determines an optimal subset of model states whose dynamics should be corrected by the ANN by examining the correlation between each state and outputs through relative order. Second, an L2-regularized least-squares problem is solved to infer values of the correction terms that are necessary to minimize the discrepancy between the model predictions and available measurements. Third, an ANN is developed to generalize relationships between the values of the correction terms and the system dynamics. Lastly, the original first-principle model is coupled with the developed ANN to finalize the hybrid model development so that the model will possess generalized prediction capabilities while retaining the model interpretability. We have successfully validated the proposed methodology with two case studies, simplified apoptosis and lipopolysaccharide-induced NFκB signaling pathways, to develop hybrid models with in silico and in vitro measurements, respectively. An intracellular signaling pathway is often represented by a set of nonlinear ordinary differential equations, which translate our current knowledge about the signaling pathway into a testable mathematical model. However, predictions from such models are often subject to high uncertainty since many signaling pathways are only partially known beforehand. In this study, we propose a systematic approach to develop a hybrid model to improve model accuracy by combining machine learning and the first-principle modeling. Specifically, model correction terms are learned from discrepancy between model predictions and measurements, and these terms are added to the first-principle model to enhance the prediction accuracy. Once these correction terms are learned from the data, an artificial neural network (ANN) model is developed to find an empirical relation between the model and the correction terms so that the developed ANN can be used to posses improved predictive capabilities even in new operating conditions (i.e., generalizability). The final hybrid model is then constructed by coupling the first-principle model with the developed ANN.
Collapse
Affiliation(s)
- Dongheon Lee
- Artie McFerrin Department of Chemical Engineering, Texas A&M University, College Station, Texas, USA
- Texas A&M Energy Institute, Texas A&M University, College Station, Texas, USA
| | - Arul Jayaraman
- Artie McFerrin Department of Chemical Engineering, Texas A&M University, College Station, Texas, USA
- Department of Biomedical Engineering, Texas A&M University, College Station, Texas, USA
| | - Joseph S. Kwon
- Artie McFerrin Department of Chemical Engineering, Texas A&M University, College Station, Texas, USA
- Texas A&M Energy Institute, Texas A&M University, College Station, Texas, USA
- * E-mail:
| |
Collapse
|
9
|
Simon LL, Pataki H, Marosi G, Meemken F, Hungerbühler K, Baiker A, Tummala S, Glennon B, Kuentz M, Steele G, Kramer HJM, Rydzak JW, Chen Z, Morris J, Kjell F, Singh R, Gani R, Gernaey KV, Louhi-Kultanen M, O’Reilly J, Sandler N, Antikainen O, Yliruusi J, Frohberg P, Ulrich J, Braatz RD, Leyssens T, von Stosch M, Oliveira R, Tan RBH, Wu H, Khan M, O’Grady D, Pandey A, Westra R, Delle-Case E, Pape D, Angelosante D, Maret Y, Steiger O, Lenner M, Abbou-Oucherif K, Nagy ZK, Litster JD, Kamaraju VK, Chiu MS. Assessment of Recent Process Analytical Technology (PAT) Trends: A Multiauthor Review. Org Process Res Dev 2015. [DOI: 10.1021/op500261y] [Citation(s) in RCA: 269] [Impact Index Per Article: 29.9] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/12/2022]
Affiliation(s)
| | - Hajnalka Pataki
- Department
of Organic Chemistry and Technology, Budapest University of Technology and Economics, Műegyetem rkp. 3, H-1111 Budapest, Hungary
| | - György Marosi
- Department
of Organic Chemistry and Technology, Budapest University of Technology and Economics, Műegyetem rkp. 3, H-1111 Budapest, Hungary
| | - Fabian Meemken
- Department
of Chemistry and Applied Biosciences, ETH Zürich, Vladimir-Prelog-Weg
1, 8093 Zürich, Switzerland
| | - Konrad Hungerbühler
- Department
of Chemistry and Applied Biosciences, ETH Zürich, Vladimir-Prelog-Weg
1, 8093 Zürich, Switzerland
| | - Alfons Baiker
- Department
of Chemistry and Applied Biosciences, ETH Zürich, Vladimir-Prelog-Weg
1, 8093 Zürich, Switzerland
| | - Srinivas Tummala
- Chemical
Development, Bristol-Myers Squibb Company, One Squibb Dr, New Brunswick, New Jersey 08903, United States
| | - Brian Glennon
- Synthesis
and Solid State Pharmaceutical Centre, School of Chemical and Bioprocess
Engineering, University College Dublin, Belfield, Dublin 4, Ireland
- APC Ltd, Belfield Innovation
Park, Dublin 4, Ireland
| | - Martin Kuentz
- School of Life
Sciences, Institute of Pharma Technology, University of Applied Sciences and Arts Northwestern Switzerland, Gründenstrasse 40, 4132 Muttenz, Switzerland
| | - Gerry Steele
- PharmaCryst Consulting
Ltd., Loughborough, Leicestershire LE11 3HN, U.K
| | - Herman J. M. Kramer
- Intensified Reaction & Separation Systems, Delft University of Technology, Leeghwaterstraat 39, 2628 CB Delft, The Netherlands
| | - James W. Rydzak
- GlaxoSmithKline Pharmaceuticals, 709 Swedeland Rd, King of
Prussia, Pennsylvania 19406, United States
| | - Zengping Chen
- State Key
Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry
and Chemical Engineering, Hunan University, Changsha, Hunan 410082, PR China
| | - Julian Morris
- Centre for Process Analytics & Control Technology, School of Chemical Engineering & Advanced Materials, Newcastle University, Newcastle upon Tyne, Tyne and Wear NE17RU, U.K
| | - Francois Kjell
- Siemens nv/sa,
Industry
Automation − SIPAT Industry Software, Marie Curie Square 30, 1070 Brussels, Belgium
| | - Ravendra Singh
- CAPEC-PROCESS,
Department of Chemical and Biochemical Engineering, Technical University of Denmark (DTU), Building 229, DK-2800 Lyngby, Denmark
| | - Rafiqul Gani
- CAPEC-PROCESS,
Department of Chemical and Biochemical Engineering, Technical University of Denmark (DTU), Building 229, DK-2800 Lyngby, Denmark
| | - Krist V. Gernaey
- CAPEC-PROCESS,
Department of Chemical and Biochemical Engineering, Technical University of Denmark (DTU), Building 229, DK-2800 Lyngby, Denmark
| | - Marjatta Louhi-Kultanen
- Department
of Chemical Technology, Lappeenranta University of Technology, P.O. Box 20, FI-53851 Lappeenranta, Finland
| | - John O’Reilly
- Roche Ireland
Limited, Clarecastle, Co. Clare, Ireland
| | - Niklas Sandler
- Pharmaceutical
Sciences Laboratory, Department of Biosciences, Abo Akademi University, Artillerigatan 6, 20520 Turku, Finland
| | - Osmo Antikainen
- Division
of Pharmaceutical Technology, Faculty of Pharmacy, University of Helsinki, Yliopistonkatu 4, 00100 Helsinki, Finland
| | - Jouko Yliruusi
- Division
of Pharmaceutical Technology, Faculty of Pharmacy, University of Helsinki, Yliopistonkatu 4, 00100 Helsinki, Finland
| | - Patrick Frohberg
- Center of
Engineering Science, Thermal Process Engineering, Martin Luther University Halle-Wittenberg, D-06099 Halle (Saale), Germany
| | - Joachim Ulrich
- Center of
Engineering Science, Thermal Process Engineering, Martin Luther University Halle-Wittenberg, D-06099 Halle (Saale), Germany
| | - Richard D. Braatz
- Massachusetts Institute
of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139, United States
| | - Tom Leyssens
- Institute
of Condensed Matter and Nanosciences, Université Catholique de Louvain, Place Louis Pasteur 1, 1348 Louvain-la-Neuve, Belgium
| | - Moritz von Stosch
- REQUIMTE
- Departamento de Química, Faculdade de Ciências e Tecnologia, Universidade Nova de Lisboa, 1099-085 Caparica, Portugal
- HybPAT, Caparica, Portugal
| | - Rui Oliveira
- REQUIMTE
- Departamento de Química, Faculdade de Ciências e Tecnologia, Universidade Nova de Lisboa, 1099-085 Caparica, Portugal
- HybPAT, Caparica, Portugal
| | - Reginald B. H. Tan
- Institute
of Chemical and Engineering Sciences, A*Star, 1 Pesek Road, Singapore 627833
- Department of Chemical and Biomolecular Engineering, National University of Singapore, 4 Engineering Drive 4, Singapore 117576
| | - Huiquan Wu
- Division
of Product Quality Research, Office of Testing and Research, Office
of Pharmaceutical Science, Center for Drug Evaluation and Research, US Food and Drug Administration (FDA), Silver Spring, Maryland 20993, United States
| | - Mansoor Khan
- Division
of Product Quality Research, Office of Testing and Research, Office
of Pharmaceutical Science, Center for Drug Evaluation and Research, US Food and Drug Administration (FDA), Silver Spring, Maryland 20993, United States
| | - Des O’Grady
- Mettler Toledo
AutoChem, 7075 Samuel Morse Drive, Columbia, Maryland 20146, United States
| | - Anjan Pandey
- Mettler Toledo
AutoChem, 7075 Samuel Morse Drive, Columbia, Maryland 20146, United States
| | - Remko Westra
- FMC Technologies B.V., Delta 101, 6825 MN Arnhem, The Netherlands
| | - Emmanuel Delle-Case
- University of Tulsa, 800 South Tucker
Drive, Tulsa, Oklahoma 74104, United States
| | - Detlef Pape
- ABB Corporate Research Center, Segelhofstrasse
1K, 5405, Dättwil, Baden, Switzerland
| | - Daniele Angelosante
- ABB Corporate Research Center, Segelhofstrasse
1K, 5405, Dättwil, Baden, Switzerland
| | - Yannick Maret
- ABB Corporate Research Center, Segelhofstrasse
1K, 5405, Dättwil, Baden, Switzerland
| | - Olivier Steiger
- ABB Corporate Research Center, Segelhofstrasse
1K, 5405, Dättwil, Baden, Switzerland
| | - Miklós Lenner
- ABB Corporate Research Center, Segelhofstrasse
1K, 5405, Dättwil, Baden, Switzerland
| | - Kaoutar Abbou-Oucherif
- School of
Chemical Engineering, Purdue University, West Lafayette, Indiana 47906, United States
| | - Zoltan K. Nagy
- School of
Chemical Engineering, Purdue University, West Lafayette, Indiana 47906, United States
- Chemical
Engineering Department, Loughborough University, Loughborough, LE11 3TU, U.K
| | - James D. Litster
- School of
Chemical Engineering, Purdue University, West Lafayette, Indiana 47906, United States
| | - Vamsi Krishna Kamaraju
- Synthesis
and Solid State Pharmaceutical Centre, School of Chemical and Bioprocess
Engineering, University College Dublin, Belfield, Dublin 4, Ireland
- Department of Chemical and Biomolecular Engineering, National University of Singapore, 4 Engineering Drive 4, Singapore 117576
| | - Min-Sen Chiu
- Department of Chemical and Biomolecular Engineering, National University of Singapore, 4 Engineering Drive 4, Singapore 117576
| |
Collapse
|